ERIC Number: ED185099
Record Type: RIE
Publication Date: 1978-May
Reference Count: 0
Research Methodologies Pertinent to the Study of Schooling Effects: A Synthesis.
Rim, Eui-Do; Coller, Alan R.
A broad spectrum of methodologies pertinent to studies of schooling effects are reviewed. Methodological issues and problems are addressed according to a three-dimensional conceptual framework: (1) indicators of schooling effects; (2) study approaches; and (3) units of analysis. Problems and uses of status attainment and difference scores as indicators of schooling effects are discussed first. Study approaches to schooling effects are divided into two broad categories: experimental and nonexperimental. Methodological issues related to the experimental approaches are discussed in relation to two types of designs: experimental group only designs and control group designs. Problems related to the nonexperimental approaches are reviewed according to: partitioning of explained variance, comparison of regression coefficients, nonlinear regression methods, and causal models. Issues and problems related to the units of analysis are presented by contrasting two positions: that data should be analyzed at the individual student level, and that data should be analyzed at the classroom, school, or district level. A third position is reviewed: that multilevel analyses should be performed because schooling effects might result from many sources at many different levels. Finally, some methodological trends are identified and their implications for further schooling briefly considered. (Author/CTM)
Descriptors: Analysis of Covariance, Analysis of Variance, Correlation, Educational Research, Elementary Secondary Education, Mathematical Models, Multiple Regression Analysis, Multivariate Analysis, Outcomes of Education, Quasiexperimental Design, Research Methodology, Sampling, School Role, Statistical Analysis, Student Characteristics, Teacher Characteristics
Research for Better Schools, Inc., 444 North Third St., Philadelphia, PA 19123 ($7.00).
Publication Type: Information Analyses
Education Level: N/A
Sponsor: National Inst. of Education (DHEW), Washington, DC.
Authoring Institution: Research for Better Schools, Inc., Philadelphia, PA.
Identifiers: Coleman Report; Linear Models; Regression